GSA chip

[Aniara]

So, does anyone know if we have to re-test on the new chip?

[khazaria]

dtvmcdonald remarked: "I checked and there are about 211,000 actually useful SNPs in common between old old and new FTDNA chips."

BroderTuck observed that "It is true that the new GSA chip tests a very different set of SNPs compared to the old Omniexpress one, but
1) More or less all testing companies are now using the GSA chip, and the percentage of people tested with the new chip will continue to grow.
2) FTDNA has done something that (as far as I know) none of the other testing companies have done, namely using the 50K customizable SNPs on the GSA chip to fill in as many gaps as possible to improve compatibility between chips."

[georgian1950]

There goes the neighborhood.

[khazaria]

I just read a post in another forum by LukaszM dated April 27, 2019 in which he says "FTDNA started to impute raw data!
Someone just told me his brand new kit from FTDNA has only about 70 000 snps in Gedmatch...
He uploaded few times to check if this is true..."

[georgian1950]

It's just the new kits (in my experience).

Thank you. I hope that is the case. I'd better get moving with my research before FTDNA changes its mind.

[Ann Turner]

If you are downloading the raw data from an old FF kit, does FTDNA give it to you in the new format? Or is it just the new kits which have the new format?

It's just the new kits (in my experience).

[Ann Turner]

I just confirmed that Wilhelm HO's DNA Kit Studio tool will create a file that can be uploaded to GEDmatch Genesis. It retains lines with chromosome "0", so GEDmatch generates lots of error messages (532). But just let in run and it get there eventually.



https://wilhelmhgenealogy.wordpress.com/dna-kit-studio/



This also applies to using Excel. You can simply load the file, sort by chromosome and position, and save it (with a new file name just to be safe).

[georgian1950]

If you are downloading the raw data from an old FF kit, does FTDNA give it to you in the new format? Or is it just the new kits which have the new format?

[Ann Turner]

I tried reformatting this raw data file for another person, but it seems to still be rejected by Gedmatch Genesis, flags it with a warning of possible problem with kit.
I too used excel to reformat, sorting by chromosome, then each chromosome by position number smallest to largest.
I also removed all the zero chromosome entries, mtDNA snps and XY snps.
Saved file as a .csv and zipped as a .qz file.

Did the file you modified have any issues once uploaded to Gedmatch Genesis?

That sounds much like what I did on my second pass, although I only removed the chromosome "0" and saved the results as a text file. I zipped it with the built-in Windows "Send to Compressed (zipped) folder." It went through matching, and I compared it with kits from various sources. My first pass removed all the "irregular" SNPs, e.g those with a gene name, but that wasn't necessary.

The SNP overlap as reported by GEDmatch (which doesn't use all data) is in these groupings:

478,876 self-to-self (new FTDNA GSA)
431,149 LivingDNA v1 (GSA "Orion")
420,707 23andMe v5 (GSA)
---
341,447 LivingDNA v2 (Affymetrix "Sirius")
---
203,133 FTDNA v1 (off the shelf Illumina)
214,455 23andMe v3 (standard Illumina + 250K custom SNP)
202,430 Ancestry v1 (off the shelf Illumina)
---
175,138 Ancestry v2 (custom)
130,570 23andMe v4 (custom)

[prairielad]

For those interested, following is the new FTDNA GSA raw data format, Results values removed
https://1drv.ms/u/s!Al27wnXopRKxj3iz8idY-y8He9ql

[prairielad]

I have seen a raw data file for the new GSA chip from someone who tried and failed to upload it to GEDmatch Genesis. It has a lot of irregularities and it is sorted by rsid instead of chromosome and location. I was able to fix the file with Excel formulas, but I really think FTDNA should provide a more conventional file.

I tried reformatting this raw data file for another person, but it seems to still be rejected by Gedmatch Genesis, flags it with a warning of possible problem with kit.
I too used excel to reformat, sorting by chromosome, then each chromosome by position number smallest to largest.
I also removed all the zero chromosome entries, mtDNA snps and XY snps.
Saved file as a .csv and zipped as a .qz file.

Did the file you modified have any issues once uploaded to Gedmatch Genesis?

[John McCoy]

Absolutely! If FTDNA is going to provide raw data to the customer, it needs to be in a usable format. Sorting by rsid is clearly not helpful. There is essentially zero cost to FTDNA to select a reasonable sort order and format.

[Ann Turner]

I have seen a raw data file for the new GSA chip from someone who tried and failed to upload it to GEDmatch Genesis. It has a lot of irregularities and it is sorted by rsid instead of chromosome and location. I was able to fix the file with Excel formulas, but I really think FTDNA should provide a more conventional file.

[Parameswara]

Someone wrote in Facebook group that the new RAW data for Family Finder only contain about 630K SNPs. Does it mean FTDNA Family Finder already switch to GSA chip ?

[Slam]

FTDNA is useless when it comes to giving their paying customers current information on what is going on.

Recently FTDNA changed their raw data file format.

Is there a new raw data file format because they started using a new testing chip?

Or is there a new raw data file format because they like to mess with their paying customers?

At some point in the future the actual information will be made available to the great unwashed.

From the information I have gleamed from crypto/undercover employee's on a certain FB group, its in preparation for the switch to the new chip.

I wish I have known this before buying an additional 10 autosomal kits, which now render them useless for my plans to create Lazarus kits.